Automatic Bone Contour Identification in Medical Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current medical imaging techniques for bone identification in surgical procedures, such as orthopedic surgery, are often inaccurate and require human intervention, leading to errors and increased failure rates, especially in complex areas like the shoulder joint.
Innovation Solution
A method for automatically identifying bone contours from medical imaging data using image filtering, morphological analysis, and label association to classify shapes without human intervention, allowing for reliable and efficient bone contour recognition and surface analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional medical imaging techniques (CT scan or MRI) are used to obtain three-dimensional modeling techniques for bone contours, then practitioners can access 3D bone models, but human intervention is required to identify which bones correspond to which contours in each image, leading to approximate assessment and inaccuracies
Solution Approach 1:
The system performs automatic bone identification and contour extraction without requiring practitioner intervention. The computer program product autonomously processes medical images to identify bones and their contours, eliminating the need for manual assessment and reducing human error in bone identification.
Solution Approach 2:
The patent replaces the manual mechanical process of bone identification with an automated computer-based image processing system. The system uses algorithms to automatically detect and classify bone contours in medical images, substituting human visual inspection and manual contour drawing with automated computational methods.
2Reliability
If surgical navigation techniques are used for orthopedic surgery operations, then surgical guidance can be provided, but the techniques do not provide satisfactory results for orthopedic surgery operations aimed at placing a prosthetic implant
Solution Approach 1:
The system changes the parameters and approach of surgical navigation specifically for orthopedic implant surgery. It provides automated bone identification, contour extraction, and 3D modeling tailored to the requirements of prosthetic implant placement, adapting the navigation technique to the specific needs of orthopedic surgery rather than using generic navigation approaches.
Solution Approach 2:
The system performs preliminary automated identification and modeling of bone contours before the surgical operation. By pre-processing the medical images to accurately identify bones and extract contours, the system prepares reliable 3D models that guide the orthopedic surgery, ensuring the navigation is optimized for implant placement from the outset.
3Ease of operation
If support engineer involvement or specific training is provided for surgical navigation tasks, then the system can be operated, but this is very expensive or difficult to implement in practice
Solution Approach 1:
The system is designed to be self-sufficient and automatically performs bone identification and contour extraction without requiring operator intervention or specialized knowledge. The computer program product autonomously processes images and generates 3D models, making the system easy to operate without needing trained personnel or support engineers.
Solution Approach 2:
The patent extracts and automates the complex tasks of bone identification and contour analysis, removing these difficult operations from the practitioner's responsibilities. By automating these complex image processing tasks, the system simplifies the overall surgical navigation process and eliminates the need for specialized training in these specific technical areas.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a method of automatic identification of the contours of at least one portion of a predefined bone on the basis of a plurality of images representing parallel sections through a measurement volume comprising said portion of bone and which are obtained by a medical imaging technique. According to the invention, such a method comprises in particular: a step of obtaining at least one shape of closed contour in at least one of said filtered images; a step of associating with each of said shapes a tag selected within a predefined bone-related nomenclature; a step of classifying said shapes so as to form at least one group of shapes delimiting a common volume isolated in space; a step of selecting the shapes of said group of target shapes.